Early Stage Detection of Lung Cancer Using Ensemble Classification Techniques

Abstract

Lung cancer is a critical global health concern, accounting for a significant proportion of cancer-related deaths due to late-stage diagnosis and limited access to timely healthcare. Early detection through intelligent systems can play a pivotal role in improving survival rates. This paper presents the design and development of a web-based lung cancer stage prediction system that integrates machine learning algorithms with a user-centric interface tailored for both patients and healthcare professionals. The system employs a Random Forest classifier trained on a comprehensive dataset containing patient demographic details and clinical parameters such as smoking habits, chest pain, shortness of breath, and wheezing.Patients can sign up, verify their identity through OTP, and submit diagnostic data, upon which the system predicts the stage of lung cancer and displays the results. Doctors can log in to view patient records in both tabular and graphical formats, enabling them to monitor and analyze multiple cases effectively. The platform also offers data visualization functionalities that help users explore trends and correlations within the dataset based on features like gender, age, and exposure history.Experimental results indicate that the Random Forest model achieved a classification accuracy of 96%, showcasing its effectiveness in medical decision support. The integration of machine learning with interactive web interfaces not only enhances the diagnostic process but also facilitates remote patient management. This system demonstrates strong potential in assisting early-stage lung cancer detection and offers a scalable solution for deployment in telemedicine and clinical environments.